THE ROC CURVE METHOD FOR THE ASSESSMENT OF DIAGNOSTIC ACCURACY

This paper reviews a methodology for the assessment of diagnostic accuracy by use of the area under the receiver operating characteristic (ROC) curve. Estimation and hypothesis testing of diagnostic accuracy based on a single sample from a single diagnostic test will be discussed with examples of computation. Sample size requirements for both estimation and hypothesis testing will be discussed as well. In addition, we present a SAS ® Program with a SAS ® Macro which calculates the sample size for the estimation of diagnostic accuracy of a single diagnostic test